Direct conversion of ethylene to propylene over Ni- and W-based catalysts: An unprecedented behaviour

Article abstract of DOI:10.1016/j.catcom.2020.106091

A new ethylene to propylene catalytic process based on cascade reactions, with two catalysts and two reactors, has been explored in this study. In the first reactor loaded with Ni-AlKIT-6 catalyst, at 60–120 °C and 3 MPa, part of ethylene has been selectively transformed into 2-butene. In the second reactor, loaded with WO_x/KIT-6 catalyst, 2-butene reacted with the unconverted ethylene at 450 °C and 0.1 MPa. Unprecedented results in terms of catalytic activity, selectivity towards the formation of propylene and stability against deactivation were achieved under these conditions. More precisely, an ethylene conversion of ~85% and a selectivity to propylene of 55% and butenes of 30% remained unchanged during 24 h on reaction stream.

Full text of DOI:10.1016/j.catcom.2020.106091

Journal Pre-proof
Direct conversion of ethylene to propylene over Ni- and W-based
catalysts: An unprecedented behaviour
Rémi Beucher, Claudia Cammarano, Enrique Rodríguez-
Castellón, Vasile Hulea
PII:
S1566-7367(20)30167-9
DOI:
https://doi.org/10.1016/j.catcom.2020.106091
CATCOM 106091
Reference:
To appear in:
Catalysis Communications
Received date:
Revised date:
Accepted date:
8 May 2020
15 June 2020
17 June 2020
Please cite this article as: R. Beucher, C. Cammarano, E. Rodríguez-Castellón, et al.,
Direct conversion of ethylene to propylene over Ni- and W-based catalysts: An
unprecedented behaviour, Catalysis Communications (2020), https://doi.org/10.1016/
j.catcom.2020.106091
This is a PDF file of an article that has undergone enhancements after acceptance, such
as the addition of a cover page and metadata, and formatting for readability, but it is
not yet the definitive version of record. This version will undergo additional copyediting,
typesetting and review before it is published in its final form, but we are providing this
version to give early visibility of the article. Please note that, during the production
process, errors may be discovered which could affect the content, and all legal disclaimers
that apply to the journal pertain.
© 2020 Published by Elsevier.

Journal Pre-proof
Direct conversion of ethylene to propylene over Ni- and W-based catalysts:
An unprecedentedbehaviour
Rémi Beuchera, Claudia Cammaranoa, Enrique Rodríguez-Castellónb, Vasile Huleaa,*
vasile.hulea@enscm.fr
aInstitut Charles Gerhardt Montpellier, UMR 5253, ICGM, CNRS, Univ Montpellier,
ENSCM, Matériaux Avancés pour la Catalyse et la Santé, 34296 Montpellier, France
bDepartamento de Química Inorgánica, Facultad de Ciencias, Universidad de Málaga, 29071
Málaga, Spain
*Corresponding author.
Abstract
A new ethylene to propylene catalytic process based on cascade reactions, with two catalysts
and two reactors, has been explored in this study. In the first reactor loaded with Ni-AlKIT-6
catalyst, at 60-120 °C and 3 MPa, part of ethylene has been selectively transformed into 2-
butene. In the second reactor, loaded with WOx/KIT-6 catalyst, 2-butene reacted with the
unconverted ethylene at 450 °C and 0.1 MPa. Unprecedented results in terms of catalytic
activity, selectivity towards the formation of propylene and stability against deactivation were
achieved under these conditions. More precisely, an ethylene conversion of ~85% and a
selectivity to propylene of 55% and butenes of 30% remained unchanged during 24 h on
reaction stream.
Keywords: Tandem catalysis; Heterogeneous catalysis; Metathesis; Ethylene; Propylene; KIT-6.
1. Introduction
Propylene, a major chemical intermediate, is usually produced from fossil-based feedstock,
using highly efficient processes, such as fluidized catalytic cracking and steam cracking of
saturated hydrocarbons. Owing to the strong demand for polypropylene during the last years
(average annual demand growth of 4.5%), the supply of propylene is of great concern because
the traditional sources will not be sufficient to cover the demand. Research efforts have been
made to develop alternative routes to propylene production based on accessible raw materials,

Journal Pre-proof
such as propane, ca. propane dehydrogenation [1], methanol, ca. methanol to olefins process
[2], and ethylene.
One of the most exciting methods is the conversion of ethylene to propylene (ETP) without the
addition of other hydrocarbons [3-6]. The sustainability of this technology is guaranteed by the
fact that the ethylene production by dehydration of bioethanol has become economically
feasible [7-9]. The ETP process consists of catalytic cascade reactions, including either (a)
oligomerization/cracking or (b) dimerization/isomerization/metathesis chemical steps.
Typically, oligomerization/cracking is catalyzed by acidic microporous materials like zeolites
and SAPOs [10-12] or by multifunctional Ni-mesoporous materials [13-16], at temperatures
higher than 350 °C. Generally, mixtures of saturated and unsaturated hydrocarbons of low
selectivity to propylene, were produced in such ETP processes.
The processes based on the dimerization/isomerization/metathesis chemical steps are
operated at lower temperature (50-150 °C) in the presence of transition metal-based catalysts
[17-23]. Three consecutive reactions and three catalytic sites are involved in these processes:
(i) dimerization of ethylene to 1-butene, (ii) isomerization of 1-butene to 2-butene, and (iii)
metathesis reaction of 2-butene with unreacted ethylene to propylene, as described below.
(i) Ethylene dimerization
푠푖푡푒 1
2퐶2퐻4 → 1− 퐶4퐻8, ∆퐻2098 = −107.31푘퐽 ∙ 푚표푙−1
(ii) 1-Butene isomerization
푠푖푡푒 2
1− 퐶4퐻8 → 푡푟푎푛푠− 2 − 퐶4퐻8, ∆퐻2098 = −8.54푘퐽 ∙ 푚표푙−1
푠푖푡푒 2
1− 퐶4퐻8 →
푐푖푠− 2 − 퐶4퐻8,∆퐻2098 = −4.35 푘퐽 ∙ 푚표푙−1
(iii) Ethylene-butene metathesis
푠푖푡푒 3
퐶2퐻4 + 푡푟푎푛푠− 2 − 퐶4퐻8 → 2퐶3퐻6, ∆퐻2098 = −2.09푘퐽 ∙ 푚표푙−1
푠푖푡푒 3
퐶2퐻4 + 푐푖푠− 2 − 퐶4퐻8 → 2퐶3퐻6, ∆퐻2098 = −5.68푘퐽 ∙ 푚표푙−1
As shown in several studies [17-20], the catalytic sites involved in these reactions could coexist
on a single support. Basset and co-workers [17] used a trifunctional WH3/Al2O3 catalyst at 150
°C and 0.1 MPa for the above processes. Li et al. [18] used a single bimetallic supported
catalyst, NiSO4/Re2O7/γ-Al2O3 in an ETP process operated at 50 °C and 0.1 MPa. Catalysts like

Journal Pre-proof
MeO/Re2O7/B2O3-Al2O3 (Me = Pd or Ni) have been also designated by Buluchevskii et al. [19]
for the one-stage ETP process performed at 80 °C and 0.1-1 MPa. More recently, Ghashghaee
and Farzaneh [20] developed a nanostructured Ru-Ba-K-MgO-Al2O3 catalyst able to convert
ethylene into propylene and butenes at 70 °C and 0.1 MPa. These studies showed that using a
single catalyst, which provides different active sites, is a challenging concept. In practice, it is
very difficult to balance the catalytic functions involved in the ETP process. Consequently, over
these catalysts undesired reactions could occur, causing a decrease in propylene selectivity.
Additionally, the catalysts rapidly deactivated.
These weaknesses could be circumvented by distributing the different catalytic sites on two
catalysts. Our group developed an original ETP process based on two robust and highly active
solid catalysts [21, 22]. In a single reactor operated at 80 °C and 3 MPa, ethylene was first
converted over a Ni-AlSBA-15 catalyst into 2-butenes, which then reacted with the unconverted
ethylene over a MoO3–SiO2–Al2O3 catalyst to produce propylene. More recently, we proposed
a new efficient heterogeneous catalytic system associating Ni-AlKIT-6 and ReOx/-Al2O3 [23,
24]. Under identical mild conditions (60 °C, 3 MPa), ethylene was very selectively converted
into high added value molecules, i.e. propylene and 1-butene. Despite their remarkable initial
performance (the initial ethylene conversion was higher than 70%), the supported MoOx and
ReOx catalysts suffered from significant deactivation after 5 h on reaction stream. Working at
low temperature, the loss of catalytic activity was most probably due to the poisoning of active
centers by strongly adsorbed products.
Beside the Re- and Mo-based catalysts, W-supported oxides are also well-known metathesis
catalysts [25]. It should be noted that a WOx/SiO2 catalyst is used in industrial metathesis
processes for converting ethylene and 2-butene to propylene [26]. While ReOx and MoOx
catalysts operate at low temperatures, i.e. 20-80 °C or 80-250 °C, WOx catalyst is active only
at temperatures higher than 400 °C. The long catalyst lifetime compared to MoOx or ReOx, the
easy and efficient regeneration and the resiliency to oxygenate molecules in the feed are the
major advantages of supported WOx catalysts [27].
In the present study, we prepared and characterized two mesoporous functionalized materials,
i.e. Ni-AlKIT-6 and WOx/KIT-6. They were evaluated as catalysts in the ETP process based on
cascade reactions: dimerization/isomerization over Ni-AlKIT-6 and metathesis over WOx/KIT-
6. The catalytic sites have been probed by ammonia temperature-programmed desorption (NH3-
TPD), hydrogen temperature-programmed reduction (H2-TPR), X-photoelectron spectroscopy
(XPS) and 27Al MAS NMR.

Journal Pre-proof
KIT-6 is a mesoporous material, possessing a bicontinuous cubic structure with Ia3d symmetry
and an interpenetrating cylindrical pore system [28]. It is a suitable support for the present
catalysts, thanks to its high surface area and large interconnected mesopores, which are key
factors for high mass transport properties. As known, these properties are crucial for processes
that involve reactive olefins [3, 6].
2. Experimental
2.1. Catalysts preparation
The dimerization/isomerization Ni-AlKIT-6 catalyst has been prepared in three successive
steps, as shown in Fig. S1 (ESI). Initially, a pure siliceous material (sample KIT-6) was
prepared in water, using (EO)20(PO)70(EO)20 triblock copolymer (Pluronic P123, Mn = 5 800 g
-1
mol , Aldrich), tetraethyl orthosilicate (TEOS 99%, Aldrich), HCl 37.5% (Aldrich) and n-
butanol (99%, Alfa Aesar). The molar ratio of reagents was 1 TEOS/0.017 P123/1.9 HCl/1.3
n-butanol/195 H2O. After stirring for 24 h at 35 °C, the mixture was kept under static conditions
at 100 °C for 24 h in an autoclave. The solid was recovered, washed with water, dried at 80 °C
and calcined for 8 h in air-flow at 550 °C.
Al-containing material (sample AlKIT-6) was obtained from KIT-6 and sodium aluminate
(99%, Carlo Erba) according to the procedure developed by our group [29]. The solid (2 g of
Na-AlKIT-6) was then converted into NH4-AlKIT-6 by ionic exchange (2 h, 30 °C, and 100
mL of 0.5 M NH4NO3 aqueous solution). The sample was subjected to nickel-ion exchange
with a 0.5 M aqueous solution of Ni(NO3)2 obtained from hexahydrate nickel(II) nitrate (98%,
Alfa Aesar), following the same procedure as described above. The exchanged sample was
dried and then calcined for 5 h in air-flow at 550 °C to achieve the catalyst denoted as Ni-
AlKIT-6. A portion of NH4-AlKIT-6 was calcined for 5 h in airflow at 550 °C to obtain the H-
AlKIT-6 sample.
WOx-based metathesis catalysts, containing 4 and 8 wt% W were prepared by the incipient
wetness impregnation method using an aqueous solution of ammonium metatungstate (99%,
Honeywell) and KIT-6, Na-AlKIT-6 or H-AlKIT-6 as supports. Before impregnation, the exact
amount of water required to fill the support pores was determined. The proper amount of
ammonium metatungstate for achieving the requested tungsten loading was dissolved in an
appropriate amount of distilled water. This solution was added to the support dropwise. The
wet material was dried at 80 °C for at least 6 h and calcined in static air at 550 °C for 8 h. These

Journal Pre-proof
materials will be denoted as X-W-KIT-6, X-W-Na-AlKIT-6 and X-W-H-AlKIT-6 (where X =
4 or 8 represents the W loading, wt%).
2.2. Catalysts characterization
The powder XRD measurements were performed on a Bruker AXS D8 diffractometer. The
solid textures were evaluated by nitrogen adsorption/desorption at -196 °C, using a
Micrometrics ASAP 2020 analyzer. 27Al MAS NMR spectra of the calcinated samples were
obtained as previously reported [29] on a Varian 600 MHz WB Premium Shielded
spectrometer, equipped with 3.2 mm o.d. ZrO2 rotors at a rotation speed of 20 kHz. The fresh
and spent catalysts were characterized by X-ray photoelectron spectroscopy (XPS). The spent
catalyst, carefully recovered from the reactor was kept under a nitrogen atmosphere before
characterization. The measurements were performed on an ESCALAB 250 (Thermo Electron)
spectrometer equipped with an Al Kα source (1486.6 eV), and on a Physical Electronics
spectrometer (PHI Versa Probe II Scanning XPS Microprobe with monochromatic X-ray Al
K radiation (100 m, 100 W, 20 kV, 1486.6 eV) and a dual-beam charge neutralizer. The
procedure has been previously described [24].
The acidity of Ni-AlKIT-6 was explored by temperature-programmed desorption (TPD) using
ammonia as probe molecule (Micromeritics AUTOCHEM 2910). H2-TPR measurements were
performed using the same apparatus. The sample (~40 mg) was pre-treated for 6 h in air-flow
at 550 °C, and then cooled down to room temperature in Ar gas flow. The reduction of the
sample was carried out while heating from room temperature to 1100 °C with a ramp of 10 °C
min-1 using a certified gas mixture of 2 vol% H2 in Ar (Air Liquide) with a flow rate of 20 mL
min-1. Hydrogen concentration at the reactor outlet was measured by thermal conductivity
detector (TCD) after removing water by a cooling trap.
Nickel content in the catalyst was measured by atomic absorption spectrometry using a Varian
Spectra 220 apparatus and a source emitting monochromatic radiation with a wavelength of 232
nm. The sample (~40 mg) is dissolved in 100 mL of 2 mol L-1 sodium hydroxide solution.
Tungsten content measurements were performed on a PANanalytical Axios max device with
an X-ray source (4 kW rhodium tube). A 100 mg sample was used for analysis.
2.3. Catalytic tests
The performances of the catalysts were evaluated in a laboratory-scale unit (Fig. S2, ESI)
equipped with two stainless steel fixed-bed reactors (i.d. = 6 mm, length = 150 mm) coupled in

Journal Pre-proof
series. The mass flows of gases were monitored by thermal mass flow control valves (Brooks
5850S HP). Before each test, the catalysts (particle size ranging from 0.15 to 0.25 mm) were
activated in-situ for 8 h at 550 °C under a nitrogen gas flow. The catalytic measurements have
been realized at constant temperature and pressure, using a feed gas stream consisting of C2H4
and N2 (1:1 v/v). The products have been analyzed online by GC (Varian CP-3800, FID), using
a CP-PoraPLOT Q capillary column (25 m, 0.53 mm, 20 µm) and an auto-sampling valve.
Ethylene conversion (X) was calculated from the respective concentrations at the inlet (Ci) and
outlet (Co) of the reactor (X = 1 - Ci/ Co). Product selectivity for the molecule i was Si = y/2
× x / x, where x stands for mole fraction and y is a coefficient equal to C atoms in the
i
j
respective molecule. The amount of products confined in the catalyst during the tests was
evaluated by TGA using a NETZSCH TG 209C apparatus (20 mg catalysts, 20 mL min-1 air,
temperature ramp of 10 °C min−1 up to 800 °C).
3. Results and discussion
3.1. Characterisation of catalysts and supports
The powder XRD results showed that all samples prepared in this study were well-organized
mesoporous materials. Thus, the small-angle XRD patterns (Fig. S3) showed a diffraction peak
d211 (2θ = 0.97°) and a peak d220 (2θ = 1.12°) characteristic of the cubic Ia3d symmetry of KIT-
6 materials [28]. No signal was observed in the powder XRD patterns recorded at 2θ > 5°,
confirming the absence of crystalline NiO or WOx species (Fig. S4). Additionally, the nitrogen
adsorption-desorption isotherms were of type IV, with a sharp capillary condensation step,
characteristic of high-quality materials with homogeneous mesopores (Fig. S5, ESI). The main
textural properties, calculated from sorption data, as well as the composition and acidity of
different samples are summarized in Table 1.
The 27Al MAS NMR spectrum of Ni-AlKIT-6 (Fig. S6, ESI) shows a major signal at about 55
ppm, corresponding to the tetrahedrally-coordinated aluminum species. Such a result confirms
that the post-synthesis alumination with sodium aluminate at room temperature is an efficient
method to incorporate aluminum into the silica matrix.
Ni-AlKIT-6 had a Si/Al ratio of 11.0 and a Ni content of 2±0.2 wt% (Table 1). The
corresponding Ni/Al ratio is 0.25 (mol/mol), indicating an exchange level of NH4+ ions with
+
Ni2+ of about 50%. During the calcination at 550 °C (see the experimental section), NH4 ions
are decomposed into ammonia and H+, generating the Brönsted acidic sites associated to

Journal Pre-proof
aluminum atoms. Note that the acid sites are useful for the 1-butene isomerization in the ETP
process [21, 23].
The oxidation state of Ni and Re sites in the catalysts was examined by XPS. The deconvoluted
spectrum of the Ni-AlKIT-6 catalyst in the Ni 2p region is shown in Fig. 1. The main signal
with a binding energy of 857.7eV is attributed to isolated Ni(II) species [30], and more precisely
to tetrahedrally coordinated Ni(II). This result is confirmed by the presence of its primary
satellite peak at 863 eV. The weaker signal at 854.2 eV reveals the presence of NiO, which was
not detected by XRD because of its small amount. It is commonly accepted that only the isolated
Ni species are precursors of the catalytic dimerization/oligomerization sites [29, 30].
Fig. 2 shows the deconvoluted XP spectra of the fresh 4-W-KIT-6 and after reaction (24 h)
catalyst samples in the W 4f region. Four contributions at 35.0, 36.5, 37.2 and 38.6 eV can be
distinguished in the spectra. According to the literature, the doublet W 4f7/2-W 4f5/2 recordedat
35.0 and 37.2 eV can be assigned to W5+ oxidation state, while the second and more intense
doublet at 36.5 and 38.6 eV corresponds to W6+ [31, 32]. The XPS W 4f core level spectrum of
the spent catalyst also shows two doublets, but the relative intensity of the doublet at high
binding energy assigned to W6+ is much higher than in the case of the fresh catalyst sample.
The total acidity of W-based catalysts was evaluated by ammonia TPD. As expected, the TPD
profiles shown in Fig. 3 and the amount of adsorbed/desorbed ammonia (Table 1) strongly
depends on the composition of catalytic materials. Both Al and W generated acid sites when
added to the parent KIT-6 silica (4-W-KIT-6, Na-AlKIT-6 and 4-W-Na-AlKIT-6, Fig. 3). For
these materials, ammonia desorbed at T < 300 °C, which is typical for the weak acid sites. The
number and strength of acid sites increased when the sodium ions are replaced by protons (H-
AlKIT-6 and 4-W-H-AlKIT-6 samples). The large desorption trace up to 450 °C indicates the
presence of medium strength acid sites.
The reducibility of W species has been examined by temperature-programmed reduction with
hydrogen (H2-TPR). The TPR profiles of 4-W-KIT-6, 4-W-Na-AlKIT-6 and 4-W-H-AlKIT-6
revealed a first H2 consumption signal at 825 °C and a second one at 1000, 975 and 1025 °C,
respectively (Fig. S7, ESI). The two reduction bands can be assigned to the reduction of W6+
and W5+ isolated species into metallic W in a two-steps process: W6+/W5+ → W4+ → W0 [33,34].

Journal Pre-proof
The differences in high-temperature reduction domains could be attributed to the different
interactions between W species and the supports.
The absence of bulky WO3 was demonstrated by X-ray diffraction and H2-TPR analyses. The
latter analysis showed an interaction between the support and tungsten. W6+ and W5+ are thus
most probably found as isolated WOx species on the surface of silica. According to the literature
[25], two isolated W6+ species have been identified. The W5+ reflects probably the same species
but bearing a hydroxyl (see Scheme 1).
To summarize, all synthesized catalysts are well organized mesoporous materials with high
surface area, pore volume and pore diameter of 7.4-8 nm. Ni-AlKIT-6 contains Ni2+ and H+
sites, while metathesis catalysts contain W6+ and W5+-species.
3.2. ETP on Ni-W catalysts
In order to produce propylene from ethylene, two catalytic beds consisting of Ni-AlKIT-6 and
W-based materials were placed in two consecutive reactors. The dimerization/isomerization
reactions occurred in the first reactor in the presence of 0.2 g of Ni-AlKIT-6 operated at 60 °C
and 3 MPa. The feed consisted of 2 L h-1 of ethylene-nitrogen mixture (1/1, v/v). The choice of
these parameters was based on our previous studies [21,29,31] and dimerization experiments
performed over Ni-AlKIT-6 (results are shown in ESI).
The complete effluent gas stream from the first reactor, containing the unreacted ethylene and
the oligomers (mainly butenes), was directed to the second reactor, where the metathesis step
occurs. The second reactor contained 0.5 g of W-based catalyst and was operated at 450 °C and
0.1 MPa. In the first reactor, over Ni-AlKIT-6, ethylene was mainly converted to butenes.
Coupling the two reactors, the main product was propylene, but some amounts of pentenes and
heptenes were formed. These results indicate that the W-based catalyst also is able to catalyze
metathesis reactions involving olefins higher than C4. Typical chromatograms obtained in the
ethylene dimerization over Ni-AlKIT-6, and dimerization-metathesis over Ni-AlKIT-6 + 4-W-
KIT-6 are given in Fig. S8 (ESI).
3.2.1. Effect of support and W loading
4-W-KIT-6, 4-W-Na-AlKIT-6 and 4-W-H-AlKIT-6 metathesis catalysts with 4 wt % W have
been coupled with Ni-AlKIT-6 to perform the ETP reactions. Table 2 summarizes the most
relevant results in terms of product distribution obtained at a similar overall ethylene conversion

Journal Pre-proof
of 50%. For these tests, ethylene conversion in the first reactor was about 35%. The results
obtained on a catalytic system containing 8 wt % W (8-W-KIT-6) are also included in Table 2.
The effect of support on the amount of each ETP product is noticeable. First, the amount of
propylene (which is the main product for all catalytic systems) decreased from 59.4% over the
4-W-KIT-6 catalyst with an Al-free support to 44.0 and 36.1% over the 4-W-Na-AlKIT-6 and
4-W-H-AlKIT-6 catalysts, respectively. On the other hand, the concentration of the C4+
hydrocarbons increased in the opposite way for the three catalysts. The selectivity in (C3 + C4)
hydrocarbons considered as useful products in the ETP process, was very high for all catalysts,
but some differences are apparent. For example, 85, 79.8 and 66.7% selectivity was obtained
over 4-W-KIT-6, 4-W-Na-AlKIT-6 and 4-W-H-AlKIT-6, respectively. Additionally, the 1-
C4/2-C4 ratio in butenes significantly decreased (Table 2). In our opinion, the acidity of the
catalyst supports can be considered as the main parameter responsible for these differences.
First, the WOx species catalyse the metathesis reactions between butenes and ethylene (Scheme
2). However, at 450 °C, the acid sites of the aluminated metathesis catalysts (i.e. 4-W-Na-
AlKIT-6 and 4-W-H-AlKIT-6) can activate a series of undesirable reactions, e.g.
oligomerization, isomerization, and cracking, which involve the primary products (propylene,
1-butene).
To conclude, in the ETP process carried out under the present conditions and with metathesis
catalysts based on KIT-6 materials, a non-acid pure silica is more efficient as support than the
silica-alumina supports. According to Le Roux et al. [35], the acidity of the support improved
the self-metathesis of propylene with a tungsten catalyst. In our study, this effect is manifested
by an over-activity of the most acidic catalyst leading to a decrease of propylene selectivity.
To establish the role of W loading, the ETP process has been carried out in the presence of Ni-
AlKIT-6 and 8-W-KIT-6 catalysts. By comparing the catalysts with 4 and 8% W (Table 2), it
can be seen that the product distribution depends on the W amount in the catalyst. Thus, the
amount of propylene and useful molecules (C3 + C4) is decreased when W loading increased
from 4 to 8%, and that in favour of the heavier molecules. These results suggest that at a higher
concentration of W sites, the catalyst exhibits an over activity in metathesis, and thus the C3
and 1-C4 primary products are consumed in subsequent reactions with other olefins. This means
that a perfect balance between the amount of Ni and W sites in the ETP process is required to
efficiently convert ethylene to propylene.
3.2.2. Evaluation of catalyst deactivation

Journal Pre-proof
The stability against deactivation of the 4-W-KIT-6 metathesis catalyst was examined in a
long-term (24 h on reaction stream) catalytic test. To increase the total ethylene conversion, the
first reactor containing the Ni-Al-KIT-6 catalyst was operated at 120 °C and 3 MPa. To ensure
a quantity of ethylene sufficient for the metathesis reaction in the second reactor, the conversion
of ethylene in the first reactor was kept 50%. The ethylene conversion, as well as the amount
of C3 and C4 among the ETP products as a function of time on stream are given in Fig. 4.
The initial conversion of ethylene was 72%, but it increased during the first 4 h to reach a
plateau at 85%. The molar selectivity to propylene decreased progressively. A 65% initial
selectivity is noticed which drops to 51% after 5 h on reaction stream, and then remains stable.
The selectivity to butenes remains constant at 27% during 24 h on reaction stream. The
average concentration in C5, C6, C7 and C8 was 9, 4, 3 and 4 vol%, respectively. The yield in
propylene (ethylene conversion x propylene selectivity) corresponding to the plateau was
45%, while the productivity in propylene was 39 mmol g-1 h-1.
To check whether the catalysts suffered from any modification after 24 h on reaction stream,
these were analysed by powder XRD, TGA and XPS. The analysis of the catalysts
diffractograms before and after reaction (Fig. S9) indicates no major structural deterioration of
the KIT-6 support after deposition of nickel and tungsten metals. The TGA profiles of the spent
nickel-based catalyst showed a weight loss of 55% (Fig. S10, ESI). The signal derivative shows
three distinctive mass-loss peaks centred at 275, 330 and 500°C. The relatively large mass
loss on this catalyst is related to its dimerization role. Heavier products formed by
oligomerization can be trapped in the pores and this confirms the need of using a support with
an adequate pore size and pore volume. The TGA profile of the spent tungsten catalyst showed
a weight loss of about 2.5% between 450 and 500 °C, proving that under the present reaction
conditions, only a small amount of heavy carbonaceous species were adsorbed on the metathesis
catalyst. The XP spectra of the fresh and used Ni-AlKIT-6 and 4-W-KIT-6 catalysts are given
in Fig. 1 and Fig. 2, respectively. The XP spectrum of the nickel catalyst after reaction shows
little change in the surface composition of nickel species (Fig. 1). The tetrahedrally coordinated
Ni(II) species are still the majority species. However, the 854.2 eV signal (highlighting the
presence of NiO species) is higher on the spent than the fresh catalyst. Thus, there is degradation
of a small part of nickel sites.
As for the fresh sample, W6+ and W5+ species are still present on the spent catalyst (Fig. 2),
but their ratio (W6+/W5+) appears increased compared to that before the reaction, ca. 61-39

Journal Pre-proof
(at.%) before the reaction to ca. 82-18 (at%) after the reaction. We have shown that W6+ can
represent isolated WOx sites on the surface of the solid and W5+ might represent the same
species but with hydroxyl attached (Scheme 1). This decrease in the amount of W5+ can thus
reflect a decrease in the hydroxyl concentration on the surface of the tungsten-based catalyst. It
is speculated that the proportion of hydroxyl might be reduced during activation under air prior
to the reaction.
3.2.3. Comparison with the literature
As stated in the Introduction section, various catalytic systems and procedures were used for
converting ethylene into propylene without adding other hydrocarbons in the process.
Inhomogeneous results were obtained, depending on the catalyst (see Table S1, ESI). Over Ni-
-
MCM-41 and Ni-MCM-48 catalysts, the specific activity in propylene was 0.8-3.1 mmol gcatal
-1
1 h-1 at 350-400 °C [13-15]. Using W-H3/γ-Al2O3, Taoufik et al. [17] obtained 9.6 mmol gcatal
h-1 at 150 °C. Over polymetallic NiSO4/Re2O7/γ-Al2O3 [18] and MeO/Re2O7/B2O3-Al2O3 (Me
= Pd or Ni) [19] catalysts, the specific activity of propylene was 12.6 and 2.8 mmol gcatal-1 h-1,
respectively. The best results were obtained by combining two distinct dimerization/metathesis
catalysts in a single reactor: 48 mmol gcatal-1 h-1 over Ni-AlSBA-15 + MoO3–SiO2–Al2O3 [21],
28 mmol gcatal-1 h-1 over Ni-AlSiO2 + MoOx/(Al)SiO2 [22], and 26 mmol gcatal-1 h-1 over Ni-
AlKIT-6 + ReOx/γ-Al2O3 [23]. In this study, using two catalysts (Ni-AlKIT-6 + WOx/KIT-6)
-1
and two reactors, we were also able to obtain a high ethylene productivity, i.e. 40 mmol gcatal
h-1. Moreover, these catalysts showed very high stability against deactivation (Fig. 4). Note that
the catalytic systems previously evaluated in all ETP processes suffered from significant
deactivation [13-23]. Fig. 5 compares the yield in propylene vs. time on stream for the dual
catalytic system, Ni-AlKIT-6 + 4-W-KIT-6 with the yields obtained over Ni-AlSBA-15 +
MoO3–SiO2–Al2O3 [21] and Ni-AlKIT-6 + ReOx/γ-Al2O3 dual catalytic systems reported [23].
Undoubtedly, the W-based catalytic system is more efficient and much more stable than the
other ones.
The direct conversion of ethylene to propylene has been investigated using two dual
catalytic systems and two reactors in series. The cascade reactions involved in the process, i.e.
dimerization/isomerization/metathesis were catalyzed by Ni2+, H+ (both delivered by Ni-
AlKIT-6 catalyst) and W (from WOx/KIT-6 catalyst), respectively. Even if the ETP process
was governed by the metathesis step, a perfect balance between the amount of Ni and W sites

Journal Pre-proof
is essential to efficiently convert ethylene to propylene. For the W-based metathesis catalysts,
the support consisting of a non-acid pure KIT-6 silica was more efficient than the silica-alumina
supports. Unlike catalytic systems based on MoOx and ReOx, those based on WOx were highly
active and stable over a long reaction period. Thus, the ethylene conversion (~ 85%), as well as
the yield in propylene (45%) remained unchanged during 24 h on reaction stream.
The encouraging results obtained in the present work about the catalytic conversion of
ethylene into propylene could create the basis for new research focused on the conversion of
non-oil-based sources into low-carbon olefins. Thus, the ETP process could be incorporated as
a second stage into an integrated system, including the bio-ethanol dehydration to ethylene and
its conversion into propylene.
Acknowledgments
ERC thanks for financial support project RTI2018-099668-BC22 of Ministerio de Ciencia,
Innovación y Universidades and FEDER.
Credit author statement
Remi Beucher: Investigation, Writing- Original draft preparation; Claudia Cammarano:
Writing- Reviewing and Editing, Validation; Enrique Rodríguez-Castellón: Investigation,
Validation; Vasile Hulea: Conceptualization, Writing- Reviewing and Editing, Supervision
Declaration of interests
☒ The authors declare that they have no known competing financialinterests orpersonal
relationships thatcould have appeared to influence the work reported in this paper.
Supplementary data

Journal Pre-proof
Supplementary material
References
[1] A.V. Lavrenov, L.F. Saifulina, E.A. Buluchevskii, E.N. Bogdanets, Propylene production
technology: Today and tomorrow, Catal. Ind. 7 (2015) 175–187.
[2] P. Tian, Y. Wei, M. Ye, Z. Liu, Methanol to olefins (MTO): From fundamentals to
commercialization, ACS Catal. 5 (2015) 1922–1938.
[3] V. Hulea, Toward platform chemicals from bio-based ethylene: Heterogeneous catalysts
and processes, ACS Catal. 8 (2018) 3263−3279.
[4] M. Ghashghaee, Heterogeneous catalysts for gas-phase conversion of ethylene to higher
olefins, Rev. Chem. Eng. 34 (2018) 595-655.
[5] V. Blay, E. Epelde, R. Miravalles, L. Alvarado Perea, Converting olefins to propene:
Ethene to propene and olefin cracking, Catal. Rev. 60 (2018) 278-335.
[6] V. Hulea, Direct Transformation of butenes or ethylene into propylene by cascade catalytic
reactions, Catal. Sci. Technol. 9 (2019) 4466-4477.
[7] J. Sun, Y. Wang, ACS Catal. Recent advances in catalytic conversion of ethanol to
chemicals, 4 (2014) 1078-1090.
[8] M. Zhang, Y. Yu, Dehydration of ethanol to ethylene, bioethylene production from
ethanol: A review and techno‐economical evaluation, Ind. Eng. Chem. Res. 52 (2013) 9505-
9514.
[9] A. Mohsenzadeh, A. Zamani, M.J. Taherzadeh, Bioethylene production from 1st and 2nd
generation ethanol: A mini review and techno-economical evaluation, ChemBioEng Rev. 4
(2017) 1-18.
[10] H. Oikawa, Y. Shibata, K. Inazu, Y. Iwase, K. Murai, S. Hyodo, G. Kobayashi, T. Baba,
Highly selective conversion of ethene to propene over SAPO-34 as a solid acid catalyst, Appl.
Catal. A: Gen. 312 (2006) 181-185.

Journal Pre-proof
[11] W. Dai, X. Sun, B. Tan, G. Wu, L. Li, N. Guan, M. Hunger, Verifying the mechanism of
the ethene-to-propene conversion on zeolite H-SSZ-13, J. Catal. 314 (2014) 10-20.
[12] J. W. Jun, N.A. Khan, P.W. Seo, C.U. Kim, S.H. Jhung, Conversion of Y into SSZ-13
zeolites and ethylene-to-propylene reactions over the obtained SSZ-13 zeolites, Chem. Eng. J.
303 (2016) 667-674.
[13] M. Iwamoto, Y. Kosugi, Selective conversion of ethylene to propene and butenes on nickel
ion-loaded mesoporous silica catalysts, J. Phys. Chem. C. 111 (2007) 13−15.
[14] A. Frey, O. Hinrichsen, Comparison of differently synthesis Ni(Al)MCM-48 catalysts in
the ethene to propene reaction, Microporous Mesoporous Mater. 164 (2012) 164−171.
[15] L.A. Perea, T. Wolff, P. Veit, L. Hilfert, F.T. Edelmann, C. Hamel, A. Seidel-Morgenstern,
Alumino-mesostructured Ni catalyst for the direct conversion of ethene to propene, J. Catal.
305 (2013) 154−168.
[16] M. Stoyanova, U. Bentrup, H. Atia, E.V. Kondratenko, D. Linke, U. Rodemerck, The role
of speciation of Ni2+ and its interaction with the support for selectivity and stability in the
conversion of ethylene to propene, Catal. Sci. Technol. 9 (2019) 3137–3148.
[17] M. Taoufik, E. Le Roux, J. Thivolle-Cazat, J. M. Basset, Direct transformation of ethylene
into propylene catalyzed by a tungsten hydride supported on alumina: Trifunctional single-site
catalysis, Angew. Chem., Int. Ed. 46 (2007) 7202−7205.
[18] L. Li, R.D. Palcheva, K.J. Jens, Conversion of ethene to propene by a dual function
NiSO4/Re2O7/γ-Al2O3 catalyst, Top. Catal. 56 (2013) 783−788.
[19] E.A. Buluchevskii, M.S. Mikhailova, A.V. Lavrenov, One-stage synthesis of propylene
from ehylene using Pd-Re2O7/B2O3-Al2O3 catalyst, Chem. Sustainable Dev. 21 (2013) 47−51.
[20] M. Ghashghaee, V. Farzaneh, Nanostructured hydrotalcite-supported RuBaK catalyst for
direct conversion of ethylene to propylene, Russ. J. Appl. Chem. 91 (2018) 972-976.
[21] R.D. Andrei, M.I. Popa, F. Fajula, C. Cammarano, A. Al Khudhair, K. Bouchmella, H.
Mutin, V. Hulea, Ethylene to propylene by one-pot catalytic cascade reactions, ACS Catal.
5 (2015) 2774−2777.
[22] R.D. Andrei, M.I. Popa, C. Cammarano, V. Hulea, Nickel and molybdenum containing
mesoporous catalysts for ehylene oligomerization and metathesis, New J. Chem. 40 (2016)
4146-4152.

Journal Pre-proof
[23] R. Beucher, R.D. Andrei, C. Cammarano, A. Galarneau, F. Fajula, V. Hulea, Selective
production of propylene and 1-butene from ethylene by catalytic cascade reactions, ACS Catal.
8 (2018) 3636-3640.
[24] R. Beucher, C. Cammarano, E. Rodríguez-Castellón, V. Hulea, Direct transformation of
ethylene to propylene by cascade catalytic reactions under very mild conditions, Ind. Eng.
Chem. Res. 59 (2020) 7438−7446.
[25] S. Lwin, I.E. Wachs, Olefin metathesis by supported metal oxide catalysts, ACS Catal. 4
(2014) 2505−2520.
[26] J.C. Mol, Industrial applications of olefin metathesis, J. Mol. Catal. A.: Chem. 213
(2004) 39−45.
[27] S. Lwin, Y. Li, A.I. Frenkel, I.E. Wachs, Nature of WOx sites on SiO2 and their
molecular structure–reactivity/selectivity relationships for propylene metathesis, ACS Catal. 6
(2016) 3061−3071.
[28] F. Kleitz, S.H. Choi, R. Ryoo, Cubic Ia3d large mesoporous silica: synthesis and
replication to platinum nanowires, carbon nanorods and carbon nanotubes, Chem. Commun.
(2003) 2136-2137.
[29] R.D. Andrei, M.I. Popa, F. Fajula, V. Hulea, Heterogeneous oligomerization of ethylene
over highly active and stable Ni-AlSBA-15 mesoporous catalysts, J. Catal. 323 (2015) 76-84.
[30] S. Moussa, P. Concepción, M.A. Arribas, A. Martínez, Nature of active nickel sites and
initiation mechanism for ethylene oligomerization on heterogeneous Ni-beta catalysts, ACS
Catal. 8 (2018) 3903-3912.
[31] M. Lallemand, A. Finiels, F. Fajula, V. Hulea, Continuous stirred tank reactor for ethylene
oligomerization catalyzed by NiMCM-41, Chem. Eng. J. 172 (2011) 1078-1082.
[32] M. Seifollahi Bazarjani, M. Hojamberdiev, K. Morita, G. Zhu, G. Cherkashinin, C. Fasel,
T. Herrmann, H. Breitzke, A. Gurlo, R. Riedel, Visible light photocatalysis with c-WO3-
x/WO3×H2O nanoheterostructures in situ formed in mesoporous polycarbosilane-siloxane
polymer, J. Am. Chem. Soc. 135 (2013) 4467-4475.

Journal Pre-proof
[33] D. Vermaire, P. Van Berge, The preparation of WO3TiO2 and WO3Al2O3 and
characterization by temperature-programmed reduction, J. Catal. 116 (1989) 309-317.
[34] R. Kourieh, S. Bennici, M. Marzo, A. Gervasini, A. Auroux, Investigation of the
WO3/ZrO2 surface acidic properties for the aqueous hydrolysis of cellobiose, Catal. Commun.
19 (2012) 119-126.
[35] E. Le Roux, M. Taoufik, A. Baudouin, C. Copéret, J. Thivolle-Cazat, J.M. Basset, B.M.
Maunders, G.J. Sunleyc, Silica‐alumina‐supported, tungsten‐based heterogeneous alkane
metathesis catalyst: Is it closer to a silica‐ or an alumina‐supported system?, Adv. Synth.
Catal. 349 (2007) 231-37.
Fig. 1. Ni 2p3/2 X-ray photoelectron spectra recorded on Ni-AlKIT-6 sample before (fresh) and
after 24 h of reaction (aged).
Fig. 2. W 4f X-ray photoelectron spectra of 4-W-KIT-6 sample before (fresh) and after 24 h of
reaction (aged).
Fig. 3. Ammonia TPD profiles of the supports alone and the various catalysts.
Fig. 4. Ethylene conversion (○), C3 (□) and C4 (◊) selectivity vs. time on reaction stream.
Reaction conditions: Feed = 2 L h-1 of ethylene-nitrogen mixture (1/1, v/v);
Dimerization reactor: Ni-AlKIT-6 (0.5 g), T = 120 °C, P = 3.0 MPa; Metathesis reactor: 4-W-
KIT-6 (0.5 g), T = 450 °C, P = 0.1 MPa.
Fig. 5. Yields in propylene over (□) Ni-Mo dual catalytic system - ref. 21, (Δ) Ni-Re dual
catalytic system - ref. 23, and (●) Ni-W catalysts - this study ; Yield = ethylene conversion x
selectivity in propylene.Ni-Mo and Ni-Re dual catalytic systems were evaluated in a single
fixed-bed reactor at 3.0 MPa and 80 °C or 60 °C, respectively.
Scheme 1. Isolated sites of W6+ and W5+ species on the surface of the metathesis catalysts.
Scheme 2. Primary metathesis reactions between ethylene and n-butenes.

Journal Pre-proof
Table 1. Composition, textural properties and surface acidity of catalysts.
b
b
b
Sample
Si/Al
Mea
DP
VP
SBET
Acidityc
(mol mol-1)
(wt %)
(nm)
(mL g-1)
(m2 g-1)
(mmol
NH3 g-1)
Ni-AlKIT-6
11.0±0.2
-
2±0.2
7.3
8.1
7.4
7.4
0.73
0.95
0.71
0.67
440
670
420
400
0.20
0.10
0.18
0.32
4-W-KIT-6
4.5±0.1
4.6±0.1
4.5±0.1
4-W-Na-AlKIT-6
4-W-H-AlKIT-6
11.0±0.2
11.0±0.2
a Me = Ni or W, bBdB pore diameter (DP), total pore volume (VP), BET specific surface area (SBET).
c from NH3-TPD.
Table 2. Ethylene conversion and product distribution in the ETP process: effect of support
used in the metathesis catalystsa.
Metathesis
catalyst
C2 conv.b
( %)
Product distribution (%mol)
C4 isomers b (%)
C3
C4
C5
C6
C7
C8
1-C4
81.1
57.7
41.6
65.4
2-C4
18.9
42.3
58.4
34.6
4-W-KIT-6
48.8
59.4
44.0
36.1
44.2
25.6
35.8
30.6
32.8
8.0
7.0
10.4
5.0
3.1
8.5
16.0
7.8
1.7
1.2
4.5
3.5
2.0
3.5
2.3
6.5
4-W-Na-Al-KIT-6 50.4
4-W-H-Al-KIT-6
8-W-KIT-6
56.2
48.0
a Reaction conditions: Feed = 2 L h-1 of ethylene-nitrogen mixture (1/1, v/v); First reactor: 0.2 g of Ni-
AlKIT-6, 60 °C, 3 MPa; Second reactor: 0.5 g of W-based catalyst, 450 °C, 0.1 MPa; b after the second
reactor.
Graphical abstract
Highlights

Journal Pre-proof
 This is an efficient catalytic way for converting ethylene to propylene (ETP)
 The ETP process is conducted in cascade mode, using two catalysts
 In the first reactor, over Ni-AlKIT-6, ethylene is selectively converted into n-C4
 In the second one, over WOx/KIT-6, the unconverted ethylene with the formed n-C4
will generate propylene
 Both catalysts are highly active and stable in the ETP process

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5